def generateSplitNode(self, datas, prenode, depth, maxD):
t_r = []
t_start = datas[0].now()
if depth >= maxD:
tNode = node((t_start, -1), datas)
tNode.word_type = 'Continue'
return [tNode]
t_num = {}
for data in datas:
t_next = data.nextLoc()
if t_next not in t_num:
t_num[t_next] = []
t_num[t_next].append(data)
t_v = []
for key in t_num:
if float(len(t_num[key])) / float(len(datas)) >= self.R and len(
t_num[key]) >= self.C:
t_node = node((t_start, key), t_num[key])
t_node.getNodeType()
t_node.upDataByType()
if t_start != key:
if t_node.word_type != 'F':
t_node.children = t_node.children + self.generateSplitNode(
t_num[key], t_node, depth + 1, maxD)
t_r.append(t_node)
else:
childNodes = t_node.splitNode()
for childnode in childNodes:
childnode.children = childnode.children + self.generateSplitNode(
childnode.ids, childnode, depth + 1, maxD)
t_r.append(childnode)
# t_node.children = t_node.children + self.generate_node(t_num[key])
else:
t_v.extend(t_num[key])
if len(t_v) > 0:
if len(t_v) < self.C:
t_node = node((t_start, -1), t_v)
t_node.word_type = 'LV'
t_r.append(t_node)
else:
maxData = max([m.now() for m in t_v])
print(maxData, t_start)
if maxData > t_start:
for pridata in t_v:
pridata.updateLo()
if prenode.word_type != 'LV':
t_node = node((t_start, maxData), t_v)
t_node.word_type = 'LV'
t_node.children = t_node.children + self.generateSplitNode(
t_v, t_node, depth + 1, maxD + 1)
t_r.append(t_node)
else:
t_r.extend(
self.generateSplitNode(t_v, prenode, depth + 1,
maxD))
else:
t_node = node((t_start, -1), t_v)
t_node.word_type = 'LV'
t_r.append(t_node)
return t_r
def generate_node(self, n_data):
t_r = []
t_start = n_data[0].now()
t_num = {}
for data in n_data:
nextLoc = data.nextLoc()
if nextLoc not in t_num:
t_num[nextLoc] = []
t_num[nextLoc].append(data)
t_v = []
for key in t_num:
if float(len(t_num[key])) / float(len(n_data)) >= self.R and len(
t_num[key]) >= self.C:
t_node = node((t_start, key), t_num[key])
t_node.getNodeType()
t_node.upDataByType()
if t_start != key:
t_node.children = t_node.children + self.generate_node(
t_num[key])
t_r.append(t_node)
else:
t_v.append(t_num[key])
if len(t_v) > 0:
for pridata in t_v:
pridata.updateLo()
t_node = node((t_start, -1), t_v)
t_r.append(t_node)
return t_r
def generateSplitNT(self, h=10):
t_ids = []
self.tree = node()
self.tree.word_type = 'root'
self.tree.children = self.generateSplitNode(self.datas, self.tree, 0,
h)
return self.tree
def generate_T(self):
t_ids = []
for id in self.datas:
t_ids.append(id)
self.tree = node()
self.tree.children = self.generate_node(t_ids)
return self.tree
def generate_node(self, datas):
t_r = []
t_start = datas[0].now()
t_num = {}
for data in datas:
t_next = data.nextLoc()
if t_next not in t_num:
t_num[t_next] = []
t_num[t_next].append(data)
t_v = []
for key in t_num:
if float(len(t_num[key])) / float(len(datas)) >= self.R and len(
t_num[key]) >= self.C:
t_node = node((t_start, key), t_num[key])
if t_start != key:
t_node.children = t_node.children + self.generate_node(
t_num[key])
t_r.append(t_node)
else:
t_v.extend(t_num[key])
if len(t_v) > 0:
t_node = node((t_start, -1), t_v)
t_r.append(t_node)
return t_r
def generate_T(self):
t_ids = []
self.tree = node()
self.tree.children = self.generate_node(self.datas)
return self.tree